Managing transient immune complex reactions in patients with paroxysmal nocturnal hemoglobinuria: clinical observations from the COMMODORE 1 and 2 studies

Abstract

Background:

Crovalimab is a novel C5 inhibitor that enables rapid and sustained C5 inhibition with every 4-week subcutaneous maintenance dosing, with the possibility for self-administration. When switching from another C5 inhibitor (binds to a different epitope than crovalimab) to crovalimab and vice versa, transient immune complexes will form and may cause transient immune complex reactions (TICRs).

Objectives:

To assess TICR occurrence, manifestation, and management in patients with paroxysmal nocturnal hemoglobinuria (PNH) who switched from another C5 inhibitor to crovalimab.

Design:

COMMODORE 1 and 2 randomized C5 inhibitor-experienced and -naïve patients, respectively, to receive crovalimab or eculizumab. The COMMODORE 1 nonrandomized, descriptive cohort included patients who previously received ravulizumab or approved or higher-than-approved doses of eculizumab.

Methods:

Pooled data of patients who switched from eculizumab or ravulizumab to crovalimab were evaluated for TICR incidence and severity. TICR treatments and TICR durations were assessed by severity.

Results:

This descriptive analysis included 201 patients who switched from eculizumab (n = 174) or ravulizumab (n = 27) to crovalimab. Baseline characteristics were generally balanced between patients with and without a TICR. Thirty-nine of 201 patients (19%) experienced TICRs (11% Grades 1–2; 8% Grade 3; no Grades 4–5). Median time to onset and median TICR duration were 1.6 (range, 0.7–4.4) and 1.7 weeks (range, 0.4–34.1), respectively. The most common symptoms were arthralgia (45%), rash (34%), and pyrexia (21%), with no evidence of renal manifestations. Oral corticosteroids were the most common TICR treatment. Grade 3 TICRs were treated with higher oral corticosteroid dose but did not take longer to resolve than Grades 1–2 TICRs.

Conclusion:

Pooled COMMODORE 1 and 2 data show that TICRs from switching between C5 inhibitors were generally mild to moderate and resolved with appropriate treatment. These results further confirm that crovalimab is well tolerated in patients with PNH.

Trial registration:

NCT04432584; NCT04434092.

Introduction

The terminal complement inhibitors targeting C5, eculizumab and ravulizumab, have become the standard of care for the treatment of paroxysmal nocturnal hemoglobinuria (PNH), an acquired rare hematopoietic stem cell disorder, in countries where they are available, based on data from randomized trials and over 15 years of real-world evidence.^1–6 Crovalimab is a novel, humanized, anti-C5 monoclonal antibody that is now approved for the treatment of PNH in numerous countries.^7–10

Crovalimab is a next-generation recycling antibody with novel surface-charge engineering that allows for low-volume, subcutaneous, maintenance dosing every 4 weeks following a loading series (intravenous dose on Day 1 followed by 4-weekly subcutaneous doses), with the possibility to self-administer at home.^11–13 Due to crovalimab binding to a different C5 epitope from eculizumab or ravulizumab, drug–target–drug complexes (referred to as transient immune complexes (TICs)) consisting of both antibodies and C5 will form when switching from eculizumab or ravulizumab to crovalimab, and vice versa. ¹⁴ The formation of TICs may result in transient immune complex reactions (TICRs) that typically manifest as rash, arthralgia, and/or myalgia. TICRs are essentially type III hypersensitivity (T3H) reactions, where the formation of antibody–antigen immune complexes also induces an inflammatory response.^15,16 This observation was first described in the Phase I/II COMPOSER study, which evaluated crovalimab for the treatment of patients with PNH, including patients who were switching from eculizumab.^12,14 TICRs were not observed in the 302 study when switching from eculizumab to ravulizumab since both drugs bind to the same C5 epitope, and TICs cannot form in such a setting. ³

Although both T3H reactions and TICRs involve the formation of immune complexes and manifest as symptoms affecting the skin or joints, TICRs are a new clinicopathological entity that has not been observed to have renal manifestations^12,14—the hallmark of literature-defined T3H reactions. ¹⁶ In addition, the frequency, clinical manifestation, and management of TICRs have yet to be explored in a larger patient population.

Here, we present pooled data of switch patients from the Phase III COMMODORE 1 and 2 studies,^17,18 investigating TICRs in patients who switched from either eculizumab or ravulizumab to crovalimab. Case studies on patients who experienced an additional TICR when switching back from crovalimab to another C5 inhibitor and clinical practice suggestions on the management of TICRs will also be discussed.

Methods

Background on COMMODORE 1 and 2

Details of the COMMODORE 1 and 2 study designs and eligibility criteria have been previously published.^17,18 COMMODORE 1 (NCT04432584) is a Phase III, global, randomized, open-label, multicenter, active-controlled trial that consists of two parts: (1) randomized arms evaluating crovalimab versus eculizumab in patients with PNH who had adequately controlled intravascular hemolysis receiving approved dosing of eculizumab; and (2) descriptive, nonrandomized cohorts exploring crovalimab in patients with PNH <18 years old, those previously receiving ravulizumab, those with the C5 R885H polymorphism who had poorly controlled hemolysis with eculizumab or ravulizumab, and those previously receiving eculizumab at the approved or higher-than-approved dose. ¹⁸ The Phase III, global, randomized, open-label, multicenter, active-controlled COMMODORE 2 (NCT04434092) study comprises two parts: (1) randomized arms investigating crovalimab versus eculizumab in complement inhibitor-naïve adult patients (⩾18 years old); and (2) descriptive, nonrandomized arm exploring crovalimab in pediatric patients (<18 years old). ¹⁷

Both studies had a 24-week primary treatment period. At Week 25, all patients who were initially randomized to the eculizumab arms were given the option to switch to crovalimab. All patients received crovalimab if they participated in the study extension period after Week 25.^17,18 If a patient discontinued study treatment at any time, they entered a safety follow-up period during which they could receive other treatments, including ravulizumab.

TICR monitoring and management guidelines

To minimize the risk for patients, monitoring and management suggestions were provided in the study protocols for guidance regarding patients who switched to crovalimab from other C5 inhibitors and those who discontinued crovalimab and switched to other C5 inhibitors (Supplemental Methods).^17,18 Treatment of TICRs was based on the clinical judgment of the investigator.

Objectives

The objective of this analysis was to assess the occurrence, manifestation, and management of TICRs using pooled data from the COMMODORE 1 and 2 studies. Patients who switched from eculizumab or ravulizumab to crovalimab at the start of the study (COMMODORE 1) or from eculizumab to crovalimab after the primary treatment period (after 24 weeks of treatment with eculizumab; COMMODORE 1 and 2 randomized arms) were evaluated for the incidence and severity of clinical manifestations of TICRs based on the National Cancer Institute Common Terminology Criteria for Adverse Events Version 5, with a focus on the joints, skin, and kidney. ¹⁹

Treatments used in the management of TICRs and the duration of these treatments were investigated.

Statistical analyses

The analysis population for the evaluation of TICRs comprised the C5 inhibitor-switched population, which included all enrolled patients in both COMMODORE 1 and 2 studies who switched from eculizumab or ravulizumab to crovalimab at the start of the study or from eculizumab to crovalimab after the primary treatment period (i.e., patients who were initially randomized to the eculizumab arms and switched treatment at Week 25). All TICR data were pooled across the switched population from both studies and analyzed descriptively. The time to onset of TICR was calculated from the time of first dose of study treatment to the time of onset of the first event. The TICR duration was calculated as the event end date minus the event start date plus 1 day; only resolved TICRs were considered for the calculation of the duration. Data on specific medications being recorded for TICR treatment (from baseline up to clinical cutoff date (CCOD)) and general concomitant medications of interest (within the first 30 days of switching to crovalimab) were summarized and descriptively evaluated in switched patients by TICR status and severity. Multiple uses within a specific medication class for a patient were counted only once. Dose of oral corticosteroids (in milligrams per kilogram) used for TICR treatment was standardized using corticosteroid dose equivalents^20–24 and the baseline weight of the individual patient. Additionally, an exploratory analysis was conducted to review the effect of selected concomitant medications of interest that were started before study enrollment or in the 30 days after treatment switch, for medical conditions other than PNH, on the incidence and severity of TICRs. The selection of these concomitant medications for analysis was based on their mechanism of action and potential for prophylactic activity against TICRs and included analgesics and antipyretics, antihistamines, corticosteroids, immunosuppressants, nonsteroidal anti-inflammatory drugs (NSAIDs), and opioids; medications given specifically for the treatment of a TICR were excluded from this analysis.

Results

The CCOD for this pooled TICR analysis was May 31, 2023, for both COMMODORE 1 and 2.

Patient disposition and baseline characteristics

A total of 201 patients switched from eculizumab or ravulizumab to crovalimab in these studies (COMMODORE 1, n = 133; COMMODORE 2, n = 68) and were included in this pooled analysis. Overall, 174 patients were switched from eculizumab to crovalimab, and 27 patients were switched from ravulizumab to crovalimab. Median age was 43 years (range, 13−85), including three patients (1%) aged <18 years. The median time from PNH diagnosis to enrollment was 6.4 years (range, 0.0–50.3). Before enrollment, 35% of patients had a history of aplastic anemia and 2% had myelodysplastic syndrome. Baseline characteristics were generally balanced between those who experienced a TICR (n = 39) and those who did not (n = 162) (Table 1).

Table 1.

Baseline characteristics.

Baseline characteristics	Switched from a C5i to crovalimab (n = 201)
	No TICR (n = 162)	TICR (n = 39)
Age, median (range), years	42 (13–85)	48 (22–76)
<18 years, n (%)	3 (2)	0
Sex, n (%)
Male	81 (50)	19 (49)
Female	81 (50)	20 (51)
Race or ethnicity, n (%)
Asian	70 (43)	16 (41)
Black or African American	4 (2)	2 (5)
White	80 (49)	20 (51)
Weight, median (range), kg	69.9 (44.0–126.0) a	76.0 (47.0–105.6)
⩾40 to <100 kg, n (%)	151 (94) a	36 (92)
Central LDH, mean (SD), ×ULN	1.7 (2.9)	1.1 (0.6)
Hemoglobin, median (range), g/L	104.5 (48.0–153.0)	105.0 (72.0–167.0) a
Time from PNH diagnosis to enrollment, median (range), years	6.3 (0.0–34.4) b	7.2 (0.1–50.3)
pRBC units transfused ⩽12 months prior to screening, mean (SD)	4.0 (7.9) a	2.5 (5.7) a
⩾1 unit, n (%)	70 (43) a	11 (28) a
History of PNH-relevant conditions prior to enrollment/first dose administration, n (%)
Aplastic anemia	56 (35)	14 (36)
Major vascular events	28 (17)	6 (15)
Renal impairment	25 (15)	5 (13)
Myelodysplastic syndrome	3 (2)	2 (5)

Baseline is the patient’s last observation prior to initiation of the respective study drug, that is, for patients in the eculizumab arms of COMMODORE 1 and 2 who become part of the C5 inhibitor-switch population after the primary treatment period, baseline refers to the patient’s last observation prior to initiation of crovalimab in the extension period.

^aOne patient had missing data.

^bFive patients had missing data.

C5i, C5 inhibitor; LDH, lactate dehydrogenase; PNH, paroxysmal nocturnal hemoglobinuria; pRBC, packed red blood cells; SD, standard deviation; ULN, upper limit of normal.

TIC reactions

A total of 39 patients (19%) in the COMMODORE 1 and 2 studies experienced 42 clinical manifestations of TICRs. Of these 39 patients, Grade 1 or 2 TICRs occurred in 23 (11%) patients and Grade 3 TICRs occurred in 16 (8%) patients (Table 2). No Grade 4 or 5 TICRs were observed. More details of patients with multiple reported TICR events are provided below. The median time to onset of the first TICR was 1.6 weeks (range, 0.7–4.4) after switching to crovalimab. Among patients who experienced a TICR, 59% (23 of 39) had an onset within 2 weeks and 95% (37 of 39) had an onset within 4 weeks. The median duration for the resolution of TICRs was 1.7 weeks (0.4–34.1). Grade 3 TICRs resolved with a median duration of 1.3 weeks (range, 0.4–5.3).

Table 2.

Summary of TICRs.

Summary of TICRs	Switched from a C5i to crovalimab (n = 201)
Patients with ⩾1 TICR, n (%) a	39 (19)
Grade 1 or 2	23 (11)
Grade 3	16 (8)
Total number of events, n	42
Time to first TICR onset, median (range), weeks	1.6 (0.7–4.4)
<1, n (%) b	3 (8)
⩾1 to <2, n (%) b	20 (51)
⩾2 to <4, n (%) b	14 (36)
⩾4 to <6, n (%) b	2 (5)
⩾6, n (%) b	0
Total number of resolved events, n c	37 (88)
Duration of TICRs, median (range), weeks d	1.7 (0.4–34.1)
Grade 1	1.8 (0.6–11.4)
Grade 2	2.4 (0.7–34.1)
Grade 3	1.3 (0.4–5.3)
Weeks, n (%) e
<1	7 (19)
⩾1 to <2	16 (43)
⩾2 to <4	6 (16)
⩾4 to <8	4 (11)
⩾8 to <12	2 (5)
⩾12	2 (5)

^aMultiple occurrences in an individual patient are counted once at the highest grade. No Grade 4 or 5 TICR events were observed.

^bPercentages were calculated using the total number of patients with ⩾1 TICR as the denominator; only events with available start dates were included.

^cResolved events refer to events with outcome “recovered/resolved” or “recovered/resolved with sequelae” as of clinical cutoff.

^dBased on resolved TICR events only, that is, 37 of 42 events (10 Grade 1 events; 14 Grade 2 events; 13 Grade 3 events) in 36 of 39 patients; duration of each resolved event was calculated as the event end date minus the event start date plus 1 day converted to weeks.

^ePercentages were calculated using the total number of resolved events (n = 37) as the denominator; only events with available start and end dates were included.

C5i, C5 inhibitor; TICR, transient immune complex reaction.

At CCOD, 37 of 42 TICR events were resolved, of which 23 (62%) resolved within 2 weeks and 29 (78%) resolved within 4 weeks of switching C5 inhibitors. Two of 39 patients (5%) had TICRs lasting ⩾12 weeks. One patient had symptoms of rash, neck pain, and pain in extremity (Grade 2) starting on study Day 15. The rash resolved after 15 days but neck pain and pain in extremity persisted for 239 days. The patient received paracetamol and loxoprofen, and the event resolved with no change in crovalimab treatment. The other patient had symptoms of pyrexia (Grade 1) and arthralgia (Grade 2) starting on study Day 10 and a rash on legs and feet (Grade 2) starting Day 13. Pyrexia and rash resolved after 6 and 10 days, respectively, but arthralgia persisted for 89 days. The patient received levofloxacin and ketoprofen, and the event resolved with no change in crovalimab treatment. Due to the nature of the ongoing clinical studies, five TICR events (in four patients) were ongoing at the time of CCOD for this manuscript and were recorded as unresolved (one patient had the same TICR reported as two ongoing events, i.e., T3H and axonal neuropathy).

Of the 39 patients who experienced TICRs, 38 patients had reported clinical manifestations, whereas one patient had a reported TICR without any further information (Table 3). The most frequently reported all grade signs and symptoms were arthralgia (45%), rash (34%), and pyrexia (21%), with no renal manifestation reported. Grade 3 TICR manifestations occurred in 15 patients, with arthralgia (47%), rash (20%), pyrexia (13%), myalgia (13%), and headache (13%) being the most common. One patient (3%) also reported a TICR event of Grade 3 axonal neuropathy when switching back from crovalimab to ravulizumab (see section “Patients with more than one TICR event” for further details). One patient discontinued crovalimab due to a TICR that manifested as Grade 3 vasculitic rash. TICRs did not result in any dose modifications of crovalimab.

Table 3.

Common manifestations (⩾5%) of TICRs.

Manifestations in ⩾5% of patients with TICR, n (%)	Switched from a C5i to crovalimab (n = 201)
	All gradea,b (n = 38)	Grade 1 or 2 (n = 23)	Grade 3 (n = 15)
Arthralgia	17 (45)	10 (43)	7 (47)
Rash	13 (34)	10 (43)	3 (20)
Pyrexia	8 (21)	6 (26)	2 (13)
Myalgia	6 (16)	4 (17)	2 (13)
Headache	5 (13)	3 (13)	2 (13)
Abdominal pain upper	3 (8)	2 (9)	1 (7)
Petechiae	3 (8)	2 (9)	1 (7)
Erythema	2 (5)	2 (9)	0
Nausea	2 (5)	2 (9)	0
Pain in extremity	2 (5)	2 (9)	0
Rash maculopapular	2 (5)	2 (9)	0
Asthenia	2 (5)	1 (4)	1 (7)
Fatigue	2 (5)	1 (4)	1 (7)

All patients who switched from another C5i at baseline or after the primary treatment period were clinically evaluated for manifestations of TICRs.

^aTICRs occurred in 39 patients; 38 patients had TICRs with reported clinical manifestations and one patient had a reported TICR without any further information.

^bNo Grade 4 or 5 TICR events were observed.

C5i, C5 inhibitor; TICR, transient immune complex reaction.

TICR treatments

Of the patients who experienced a TICR from baseline to CCOD (n = 39), 6 of 10 (60%) with Grade 1 TICR, 11 of 14 (79%) with Grade 2 TICR, and 15 of 15 (100%) with Grade 3 TICR received at least one treatment specifically for treating TICR (Table 4). The most common treatment for TICRs was corticosteroids, which were used in 72% of patients who received at least one TICR treatment. Of the 64 corticosteroid medications administered, 45 (70%) were given via the oral route and 9 (14%) via the topical route. Corticosteroid treatment was mainly administered via the oral and topical routes for Grade 2 TICRs and via the oral and intravenous routes for Grade 3 TICRs. A higher percentage of patients who experienced Grade 3 TICRs were treated with higher doses of oral corticosteroids (0.5 to <2 mg/kg) than patients with Grade 2 TICRs (37% and 20%, respectively). Other TICR treatments include antihistamines, analgesics and antipyretics, NSAIDs, and opioids. TICR treatment duration did not lengthen with increase in severity.

Table 4.

Summary of TICR treatments.

Summary of TICR treatments	Switched from a C5i to crovalimab and experienced ⩾1 TICR (n = 39)
	TICR Grade 1 (n = 10)	TICR Grade 2 (n = 14)	TICR Grade 3 (n = 15)
Patients with ⩾1 TICR treatment, n (%)	6 (60)	11 (79)	15 (100)
Corticosteroid	1 (10)	8 (57)	14 (93)
Antihistamine	2 (20)	5 (36)	7 (47)
Analgesic and antipyretic	0	4 (29)	6 (40)
NSAID	3 (30)	4 (29)	1 (7)
Opioid	0	1 (7)	3 (20)
Other a	1 (10)	3 (21)	5 (33)
Route of administration, n (%)
Corticosteroid	nm = 1	nm = 23	nm = 40
Intravenous	0	1 (4)	7 (18)
Oral b	0	15 (65)	30 (75)
<0.5 mg/kg	0	12 (80)	19 (63)
0.5 to <1 mg/kg	0	3 (20)	7 (23)
1 to <2 mg/kg	0	0	4 (13)
Topical	0	7 (30)	2 (5)
Transdermal	1 (100)	0	1 (3)
Antihistamine	nm = 2	nm = 6	nm = 12
Intravenous	0	0	5 (42)
Oral	2 (100)	6 (100)	7 (58)
Analgesic and antipyretic	nm = 0	nm = 9	nm = 11
Intravenous	0	0	6 (55)
Oral	0	9 (100)	5 (45)
NSAID	nm=4	nm=8	nm=1
Intravenous	0	2 (25)	0
Oral	4 (100)	6 (75)	1 (100)
Opioid	nm = 0	nm = 3	nm = 7
Intravenous	0	0	4 (57)
Oral	0	3 (100)	1 (14)
Subcutaneous	0	0	2 (29)
Other a	nm = 2	nm = 10	nm = 20
Intravenous	0	4 (40)	10 (50)
Oral	2 (100)	5 (50)	8 (40)
Subcutaneous	0	1 (10)	2 (10)
Duration of TICR treatment, median (range), days c
Corticosteroid	nm = 1 4.0 (4.0–4.0)	nm = 22 7.5 (4.0–35.0)	nm = 36 4.0 (1.0–42.0)
Antihistamine	nm = 2 3.0 (1.0–5.0)	nm = 5 8.0 (4.0–19.0)	nm = 11 3.0 (1.0–22.0)
Analgesic and antipyretic	nm = 0 NE (NE–NE)	nm = 9 8.0 (1.0–72.0)	nm = 10 1.0 (1.0–7.0)
NSAID	nm = 3 1.0 (1.0–3.0)	nm = 8 9.0 (1.0–46.0)	nm = 1 2.0 (2.0–2.0)
Opioid	nm = 0 NE (NE–NE)	nm = 3 5.0 (1.0–7.0)	nm = 7 1.0 (1.0–7.0)
Other a	nm = 2 3.0 (3.0–3.0)	nm = 10 4.0 (1.0–64.0)	nm = 20 3.5 (1.0–8.0)

n_m refers to the number of medications. Multiple uses within a specific medication class for a patient were counted only once. Patients who received more than one medication class were included in all relevant rows.

^aOther includes antinausea/vomiting, laxative, antibiotics, benzodiazepines, diuretics, low molecular weight heparin, proton pump inhibitor/other antiacid, saline/other fluids, and vitamins/minerals.

^bDose of oral corticosteroids (in milligrams per kilogram) used for TICR treatment was standardized using corticosteroid dose equivalents^20–24 and the baseline weight of the individual patient.

^cDuration was calculated as the medication end date minus the medication start date plus 1 day.

C5i, C5 inhibitor; NE, not estimable; NSAID, nonsteroidal anti-inflammatory drug; TICR, transient immune complex reaction.

Concomitant medications of interest

In an exploratory analysis of patients who switched from another C5 inhibitor to crovalimab, 36% of patients (14 of 39) with any-grade TICR and 39% (63 of 162) of those without TICR received at least one concomitant medication for conditions other than PNH. Concomitant medications were either started before study enrollment or within the first 30 days after switching treatment (Table S1). The most common concomitant medication of interest was corticosteroids, which were used in 13% of patients with any-grade TICR and 17% of patients with no TICR. Of patients who received a concomitant antihistamine (n = 17) or opioid (n = 2), none experienced a TICR. None of the patients who experienced a Grade 3 TICR received a concomitant medication of interest other than analgesics and antipyretics within the first 30 days of switching to crovalimab.

Patients with more than one TICR event

There were two patients who experienced more than one TICR. One patient received prior treatment with ravulizumab before enrollment in the nonrandomized cohort of COMMODORE 1. On Day 14 (i.e., switch Day 14 from ravulizumab), the patient experienced a TICR, manifesting as Grade 2 chills and arthralgia. The total duration of the TICR event was 12 days, and there was no change in treatment with subcutaneous crovalimab during this period. The patient permanently discontinued subcutaneous crovalimab (last dose on Day 29) due to Grade 3 treatment-related serious adverse event of sepsis and switched to treatment with ravulizumab. After switching back from crovalimab to ravulizumab (Day 52, i.e., switch Day 11), the patient experienced an additional TICR episode, which manifested as Grade 3 myalgia and axonal neuropathy (symptoms were suggestive of sensory-motor peripheral neuropathy) and Grade 2 muscular weakness and arthralgia. At CCOD, the events of axonal neuropathy and TICR were resolving.

The second patient who was randomized to the eculizumab arm of COMMODORE 2 experienced a TICR after having switched to crovalimab, following completion of the primary treatment period, which manifested as Grade 1 rash and Grade 2 edema and myalgia (Day 207, i.e., switch Day 4). The TICR lasted for 24 days and treatment with crovalimab was interrupted. The patient withdrew consent from the study (last dose of crovalimab on Day 218) and subsequently received ravulizumab, which led to a second TICR (Day 245, i.e., switch Day 2) that manifested as Grade 1 dysesthesia and Grade 2 myalgia, erythema, joint swelling, and arthralgia. The TICR event remained unresolved at CCOD.

Patients who switched from ravulizumab to crovalimab

A limited number of patients (n = 27) included in this analysis switched from ravulizumab to crovalimab. The median time to onset (2.0 weeks (range, 1.3−4.3)) and median duration of TICR (1.7 weeks (range, 0.6−11.9)) were similar to those of the overall switch population. At CCOD, all TICRs were resolved with treatment, except for the events of axonal neuropathy and TICR as described in the previous section.

The proportion of patients who experienced a TICR was 33.3% in patients who switched from ravulizumab to crovalimab compared with 17.2% in patients who switched from eculizumab to crovalimab. Results should be interpreted with caution due to the small sample size.

Suggestions on how to manage TICRs based on literature and expert opinion

With the availability of more PNH treatment options, it is important to understand the management of TICRs when switching between drugs that bind to different epitopes on the same target. There are currently no evidence-based guidelines or controlled clinical trials to form the basis of management recommendations. The TICR management algorithm developed here was adapted from available literature on serum sickness-like reactions and suggestions from the authors (Figure 1).^25–29

Figure 1.

Suggested TICR management algorithm. This algorithm was adapted from available literature on serum sickness-like reaction and suggestions from the authors.^25–29 Doses of treatment used were based on the label of each drug and literature.^26–35

^aBefore prescribing NSAIDs, please check for usual contraindications (i.e., low platelets, gastrointestinal bleeding, etc.).

BID, twice daily; IV, intravenously; NSAID, nonsteroidal anti-inflammatory drugs; PO, orally; QD, once daily; QID, four times daily; T3H, type III hypersensitivity; TICR, transient immune complex reaction.

In general, mild-to-moderate TICRs, which form the majority of these events, can be treated using oral corticosteroids, antihistamines, and NSAIDs. The approach of managing TICRs in this setting resembles the management of serum sickness reactions, with two important differences. First, the culprit agent cannot be withdrawn to stop the reaction as maintenance of terminal complement blockade is essential to avoid hemolysis. In addition, according to the lattice theory of immune complex formation, maintaining a high concentration of crovalimab favors the formation of smaller TICs, which can be eliminated from the serum at a faster rate than large TICs, making them less pathogenic. ¹⁴ This theory is supported by the transient nature and mild-to-moderate severity of TICRs observed in this analysis, with no dose modifications of crovalimab required. The second difference concerns the fever symptom, which can be part of a TICR but also a sign of an infection. As terminal complement inhibition is a known risk factor for bacterial infections, an infectious screen to rule out bacterial infection or sepsis should be conducted in the event of fever.

Discussion

In this pooled analysis of data from the Phase III COMMODORE 1 and 2 studies, TICRs that may occur when switching between different C5 inhibitor treatments ¹⁴ (i.e., from eculizumab or ravulizumab to crovalimab or vice versa in the context of this article) were observed in 19% of patients in the C5 inhibitor-switched population. These reactions showed first clinical signs and symptoms after a median of 1.6 weeks and resolved after a median of 1.7 weeks, as expected based on the transient nature of the immune complexes. Grade 3 TICRs resolved within a shorter median duration compared with Grades 1 and 2 TICRs potentially due to the more aggressive treatment with corticosteroids, leading to a quicker recovery. The lack of TICR relapse also highlights the transient nature of these events. Unlike literature-defined T3H reactions, ¹⁶ no renal manifestations were observed with TICRs. It is possible that C5 inhibition could have contributed to the lack of renal manifestations. ³⁶

No baseline characteristics distinguishing patients who experienced TICRs from those who did not experience TICRs were identified in this analysis. Predisposition factors for developing TICRs are currently being explored.

TICRs usually manifested as mild-to-moderate arthralgia and rash and the majority of TICRs resolved within 2 weeks. TICRs were generally manageable with corticosteroid, antihistamine, and NSAID treatment; only one patient discontinued from crovalimab due to a Grade 3 vasculitic rash when switching from eculizumab to crovalimab. Corticosteroids were the most common TICR treatment in the C5 inhibitor-switched population and were usually administered orally. Grade 3 TICRs were treated with higher oral corticosteroid dose but did not take longer time to resolve compared with Grade 1 or 2 TICRs.

In our exploratory analysis, no clear effect of concomitant medications (used for conditions other than PNH) was observed on the onset, duration, or severity of TICR. Notably, none of the 17 patients who received a concomitant antihistamine experienced a TICR and none of the patients with Grade 3 TICR received a concomitant medication of interest, except for analgesics and antipyretics.

Data from this analysis further confirm that crovalimab is well tolerated in patients with PNH, including those who switched from other C5 inhibitors to crovalimab. Our findings are consistent with those of the Phase I/II COMPOSER study in patients with PNH who switched from eculizumab to crovalimab. ¹⁴ These results serve to provide practical clinical guidance for the management of TICRs when switching between C5 inhibitors, the standard of care for PNH treatment in countries where they are available.

This analysis was limited by the small number of patients who switched from ravulizumab to crovalimab and those who were <18 years old. These patients were enrolled in the nonrandomized arm of COMMODORE 1, which was not designed to recruit a fixed number of patients, resulting in limited numbers. Although we observed an apparent higher proportion of patients who experienced a TICR when switching to crovalimab from ravulizumab than from eculizumab, the small sample size in the ravulizumab switch subgroup precludes the ability to draw robust conclusions on whether there are any clinically meaningful differences in TICR incidence rates between patients who switch from eculizumab to crovalimab or from ravulizumab to crovalimab. This analysis is also limited by the unavailability of paired data on the onset of TICR and breakthrough hemolysis. However, it is relevant to consider findings from a separate exploratory post hoc investigation of the COMMODORE 1 and 2 studies, which revealed that none of the observed breakthrough hemolysis events were attributed to TICR. ³⁷ These results offer some reassurance regarding a potential association between breakthrough hemolysis and TICR events, despite the limitations of our current dataset.

Conclusion

With advancements in the PNH treatment landscape and the availability of more therapeutic options, it is essential to understand more about the management of TICRs when switching between drugs that bind to different epitopes. Pooled data and case studies from the Phase III COMMODORE 1 and 2 trials show TICRs that occurred from switching between C5 inhibitors were generally mild to moderate and resolved with appropriate treatment. In this article, the authors suggest clinical practice guidance on managing these transient reactions. Overall, findings from this pooled TICR analysis further confirm that crovalimab is well tolerated in patients with PNH, supporting the potential role of crovalimab as a new PNH treatment that allows for low-volume, subcutaneous, maintenance dosing every 4 weeks, with the possibility for self-administration.